The SUN has set and you can barely see it through WINDOWS.
Lately I have been reading a lot of articles on the current happenings of Sun Microsystems. For years now, I have always had a soft spot for Sun Microsystems and despite all the struggles that they have been through, deep down I can feel that their end is approaching. If not the end, at least a complete remodel of their company will be taking place to cater to their recent investments in technology.
Now before I continue with this I wish to highlight a few statistics: recent reports have shown that year-after-year the Linux Operating System has taken a firm hold in the overall enterprise market. As of 2008 it had been holding a share of at least 13.4% (according to recent IDC reports). UNIX Operating Systems on the other hand report 32.7% usage in the same market. Combined (46.1%), Linux and UNIX out-weigh Microsoft’s share of 36.5%. I am filled with joy when I see that every year Linux increases its share by an average of 10%. With leaders such as Red Hat, Novell and Canonical, I see an even stronger future for the Linux Operating System.
Sun’s recent struggles have clearly shown the strong influence and adoption of open source software and they never displayed any issues to admit this and conform to it. While these statistics show the operating platforms within the enterprise industry it does not give us any hints on the percentage in usage of open source application running on these platforms and within these architectures.
As of 2007 (coincidently after the announcement of ZFS, dtrace and openSolaris), Sun’s stock prices and numbers have declined at a rapid rate. Revenue goals are not being matched and to help boost the usage of Sun products, more focus has been given to the open source community. Sun acquired mySQL and have been doing very well with the Open Office suite. ZFS and dtrace have been open sourced along with the Solaris Operating System. In desperation they began adopting open source so that Sun can still remain as an entity if not anything else. As long as users use mySQL or Open Office there will still be a need for Sun Microsystems. But at what cost? The focus has been dramatically shifted and emphasis has been placed on the cheaper and scalable Intel architecture as opposed to SPARC. In November of 2008 the company announced that another 5000-6000 positions will be cut. Is this giving us a glimpse of the future of Sun? Are they reshaping their company for a newer business model?
In the past couple of years their marketing has been doing nothing but driving features. Sun has barely been staying afloat because of their pushing of features and they have not held back on attacking others in the process. In the blogs of Sun employees I have seen the majority of attacks directed towards the Linux Operating System; most of which are misinforming cheap shots.
Why use anything else when you can use ZFS? Honestly, Linux has made great strides in providing enterprise class file systems/volume managers with emphasis on high availability/performance and they continue to do so. This includes device-mapper/LVM2 and the widely hyped btrfs. Note that I have seen a project to port device-mapper to openSolaris. And dtrace? Systemtap is a much worthy opponent for the Linux platform.
How long will this strategy allow them to survive in what has apparently become the jungle of Linux and Windows? Windows itself suffers from many flaws that prevent it from being a true enterprise class solution. Its I/O subsystem suffers dramatically with regards to performance and how I/O requests are handled. The NTFS file system is also one of the most horribly designed and non-scalable file systems out there. It was designed for client usage. Given time, I still see the growth of the Linux Operating System in the enterprise environment.
Can you please supply more info or a link about Window’s: “Its I/O subsystem suffers dramatically with regards to performance and how I/O requests are handled.”
Comment by xlinuks — 23. December 2008 @ 09:30
Actually, I should clarify that. The limitations actually reside in the SCSI subsystem, specifically the scatter gather lists. The purpose of the scatter gather lists is to coalesce all sequential read/write operations; upon entering the SCSI subsystem and just prior to the transfer(s) being sent to the SCSI-based media (i.e. SCSI, SAS, fibre channel disk devices, etc.). On Windows, it can be modified through a registry entry and by default it is set to 64KB. The maximum value (validated the last time I took a fibre channel trace through a protocol analyzer) despite what you enter in the registry file is 1MB. So all I/O transfers are partitioned off into a maximum of 1MB chunks and then sent to the SCSI device. The last time I checked and tested it, Linux’s scatter gather lists on the other hand can handle up to 4 MB chunk transfer (I am not sure if this is the maximum though; need to check the kernel source). Again, this is validated through a protocol analyzer. When we are dealing with larger sized I/O transfers, 1 MB tends to be a significant bottleneck. As for which registry entry controls this, it now escapes my mind but as soon as I come across it again, I will try to remember to post it.
Note that additional limitations (on both platforms) can also reside with the HBA attached to the SAN.
Comment by admin — 23. December 2008 @ 10:22
Found it! This is how I remember it: The scatter gather lists are defined in the HBA fields of the Windows registry.
HKLM\SYSTEM\CurrentControlSet\Services\[HBA-device]\Parameters\Device\
The entry is: MaximumSGList (REG_DWORD: 0xFF)
The supported values range from 16 to 255. Setting it to 255 sets the maximum of 1MB and if you go over 255 (like 256) it will default to 64KB.
Comment by admin — 23. December 2008 @ 11:21
The Sun is far from setting. Sure the organization is redefining itself, but who isn’t in this economy. There are a number of companies laying off workers. Cisco just laid off workers as well. That is business. If a company doesn’t evaluate their direction, needs and resources appropriately, they are not living up to their obligations.
Sun is providing an end to end and open integrated enterprise stack. From Hardware to Applications. Any organization that doesn’t at least look at Sun as a strategic partner for their success and to drive business value is performing a disservice to their owners and stakeholders. Their value proposition is compelling. They have innovated and continue to innovation in processors, servers, infiniband, storage, filesystems, operating systems, virtualization, databases, programming languages, application servers and frameworks, application development tools and compilers, identity and access management, SOA, HPC, Management Suites, client-side frameworsk, and thin clients.
Name me ONE organization that can say they can meet your entire IT need s and do it cost effectively (that is where IBM and HP fall out of the race :) ). Sun is leading a charge of innovation and game changing paradigms. If organizations don’t bother to at least listen, they are naive.
I use Linux as well as Solaris. And I prefer Solaris any day of the week for my applications. Much more balance, polish, and innovation. I can take advantage of the Solaris operating system’s features and capabilities to provide a competitive advantage over competing applications. The innovations of Java, ZFS, Hybrid Storage Pools, and DTrace are incredible game changers. Their hardware (new SPARC, CoolThreads, and x86) has impressive capability as well. Not to say Linux doesn’t have its moments.
When my business is at stake, I choose Solaris on Sun. It has never failed me. Even if you are a user of Linux or Windows, the Sun hardware as a hardware platform is impressive. Sun’s software is cross platform as well. Sun provides a company with CHOICE, and thus provides business agility.
Rather than take my word or this author’s word, you should find out and discover for yourself. Try it (they have a number of try and buy programs as well as developer specials), buy it, support innovation. Support companies like Sun and Juniper. Don’t be a lemming, make your own decisions.
Comment by Enlightened — 23. December 2008 @ 12:01
Thanks to the admin for the explanation.
I only regret that after inventing Java Sun almost forgotten the desktop/client side. Only after like 5 years after the applets died Sun realized there’s something bad on the web about Java and applets. That’s not the attitude of a winner. Even M$ had a faster reaction when it saw its IE market share dropping. But Sun waited until the applets died off like dinosaurs, waited a few more years, after that they started talking about support for audio/video/3D, web-designer friendliness, decent start-up time…
They do create good (perhaps best) server hardware and server-side software solutions, but they’re like newbies to what client-side customers need(ed), that’s why even they use Flash on their sites and no applets.
They learned, but not enough, 64bit plugin still not available after like 5-7 years of customers begging for it, they still require the applets to be signed for MINOR reasons and the customer has to answer weird questions about certificates and security: flash just does the job. Sun needs to understand it’s not 1996. We’re not in Kansas anymore, there is plenty of good choice in the internet now.
Comment by xlinuks — 23. December 2008 @ 18:08